Extended Meta-Analysis Fit with Voxelwise Covariate

Wrapper for meta-analysis that supports an optional voxelwise covariate. This extends the basic fmri_meta_fit to handle per-voxel covariates.

Usage

fmri_meta_fit_extended(
  Y,
  V,
  X,
  method = c("pm", "dl", "fe", "reml"),
  robust = c("none", "huber"),
  huber_c = 1.345,
  robust_iter = 2,
  voxelwise = NULL,
  center_voxelwise = TRUE,
  voxel_name = "voxel_cov",
  n_threads = getOption("fmrireg.num_threads", 0)
)

Arguments

Y

Matrix of effect sizes (S x P)

V

Matrix of variances (S x P)

X

Design matrix (S x K)

method

Meta-analysis method

robust

Robust estimation method

huber_c

Huber tuning constant

robust_iter

Number of IRLS iterations

voxelwise

Optional voxelwise covariate matrix (S x P)

center_voxelwise

Logical; center voxelwise covariate per feature

voxel_name

Name for voxelwise coefficient

n_threads

Number of threads

Value

List with meta-analysis results